Checking out checkout charity: A study of point-of-sale donation campaigns

There have been a proliferation of point-of-sale donation campaigns, where people are asked to donate following an unrelated transaction. We use an experiment to compare three popular solicitation mechanisms used in this setting: a rounding request (i.e., yes or no to an amount based on the transaction); a fixed donation request (i.e., a request to donate an amount unrelated to the transaction); and an open-ended ask. Donation rates are the highest for the rounding mechanism. Differences in donation rates between the rounding and fixed request treatments appear to be driven by “loose change effects,” whereby individuals are more likely to donate when doing so reduces the amount of change received from the coupled transaction. The open-ended ask yielded a higher per person revenue than the fixed request. However, we estimate that the average person facing the fixed request was, in fact, willing to donate more (but constrained by the amount of the ask). This suggests that the design of the fixed request campaign could be altered to yield similar or even higher revenue than open-ended. We also examine the effects of providing (limited) information on the charity, and find that this increases revenue and donation rates, but only for the fixed request mechanism. Information appears to motivate donations in situations where loose change effects are less important.